board only
Experiment E029 / terminal paired result
Two exact moves.
A measurable local gain.
Showing Luna two unranked moves from a frozen exact tablebase policy improved exact 50-move-rule utility on this synthetic five-piece population.
60 paired positions240 registered calls10-way Luna execution
Replay the registered intervention
Two exact suggestions are visible
Only the information panel changes. The move tool and legal action universe remain identical.
Unranked exact adviceSuggestion αSuggestion βNo score · no rank · any legal move allowed
Action boundaryidentical unrestricted move-string tool in both arms
Registered decision / qualified positive
+7.08 percentage points.
Board-only mean utility was 0.425; exact-top-two advice reached 0.496. The registered whole-position bootstrap interval was [2.92, 11.67] percentage points, with paired sign-flip p=0.0032399676.
Analysis provenance. The registered statistics and decision rule classify this cohort as positive, but the executable analyzer was completed after terminal data and was not hash-bound before collection.
two exact moves shown
paired advice effect
paired positions; zero provider-missing calls
Failure-preserving chronology
Seven gates.
Nothing quietly rewritten.
- 01
Original source version failed closed
NO-GO: one preregistered material-by-color outcome stratum was infeasible; zero Luna calls.
- 02
Source Selection Amendment 1
Post-tablebase-label and pre-model-outcome min-cost-flow selection preserved the feasible registered margins without generating or querying new positions.
- 03
Amended corpus qualified
The 60-position corpus has complete all-legal labels and reproduced exactly on a second provider-free parse, with a disclosed single-source limitation.
- 04
Execution interface incident
Returned action strings reached a UCI-only event boundary even though the prompt exposed UCI and SAN through an unrestricted string tool; no scored event preceded the stop.
- 05
Recovery Amendment 1 rejected
A request-body replay guard incorrectly blocked legitimate byte-identical future replicates; preflight rejected it with zero calls under the draft.
- 06
Recovery Amendment 2 accepted
Attempt identities protect against replay; legal UCI is canonicalized, otherwise only unambiguous legal SAN is converted, uniformly across both arms.
- 07
Terminal result admitted separately
A versioned content-addressed terminal aggregate is public through a separate reviewed sanitizer, with the post-run analyzer implementation-lock deviation disclosed.
The terminal analysis establishes
One local advice effect.
Exact-advice effect on 50-move-rule move utility for one Luna snapshot and one frozen synthetic five-piece population.
It still cannot establish
- full-game transfer or Elo improvement
- unassisted Luna chess strength
- base-model learning or general agentic improvement
- human-like reasoning, better vision, or better search
- benefit outside five-piece tablebase positions or for another model
Public evidence boundary
Publish the protocol.
Withhold the live game.
Included
aggregate design counts
amendment chronology
claim limits
content identities
Excluded
positions
moves
prompts
responses
request identifiers
provider metadata
live progress
tokens
costs
account state
interim outcomes
Read the terminal result, then the failure-preserving account of how the protocol reached a valid cohort without quietly erasing its amendments.